Political Parties as Drivers of U.S. Polarization: 1927-2018

Abstract

The current polarization of elites in the U.S., particularly in Congress, is frequently ascribed to the emergence of cohorts of ideologically extreme legislators replacing moderate ones. Politicians, however, do not operate as isolated agents, driven solely by their preferences. They act within organized parties, whose leaders exert control over the rank-and-file, directing support for and against policies. This paper shows that the omission of party discipline as a driver of political polarization is consequential for our understanding of this phenomenon. We present a multi-dimensional voting model and identification strategy designed to decouple the ideological preferences of lawmakers from the control exerted by their party leadership. Applying this structural framework to the U.S. Congress between 1927-2018, we find that the influence of leaders over their rank-and-file has been a growing driver of polarization in voting, particularly since the 1970s. In 2018, party discipline accounts for around 65% of the polarization in roll call voting. Our findings qualify the interpretation of -and in two important cases subvert -a number of empirical claims in the literature that measures polarization with models that lack a formal role for parties.

Ideology, Party, and Opinions of Taxes on the Rich and Middle Class in the United States: Evidence from the 2016 General Social Survey

“…Also worth noting is that moderating effects of party affiliation, Democratic and Republican, were considerably smaller in 1987 and 1996 than in 2016. This is consistent with findings of Canen, Kendall and Trebbi (2021) and Druckman, Peterson and Slothuus (2013) showing that the rapidly increasing ideological polarization of and party discipline by political elites since the 1980s and 1990s has intensified the role of party policy positions in shaping public opinion.…”

Political Parties and Hate Crimes: Empirical Evidence from the United States

“…An investigation on the effects of Legislative control can be made since it also influences public policies. The relationship between political polarization and hate crimes can also be investigated as this is a growing concern in the United States (Canen et al, 2020). Additionally, further developments for the present work can be made aggregating hate crime data to the county level, as it increases the number of observations and improves models.…”

“…Increased animosity between groups is an important factor to be observed, in which its link to politics may be seen through polarization. Increasing polarization between Republicans and Democrats has occurred during the last years, especially due to ideological polarization between the two parties (Canen et al, 2020). Moskowitz et al (2019), in addition, shows that part of this difference between parties arises from polarization in social issues, whereas economic issues are less important.…”

“…Thus, evidence shows that politicians in power, especially more extreme ones, may affect the number of hate crimes. In a context of increasing polarization, mainly driven by social issues and by Republican politicians getting more extreme than Democrats (Canen et al, 2020;Moskowitz et al, 2019), it raises the question of whether different political parties in power affect hate crimes differently. However, it is important to stress that, although recent evidence on hatred focuses on Republican politicians, this is not an issue that is only historically connected to the GOP.…”

Divided Government and Polarization: Regression-Discontinuity Evidence from US States

“…McCarty (2019) shows that most of the political elite polarization observed at both at the national and state level is due to Republicans and Democrats representing similar districts in a different way (divergence). While he suggests changes in social and economic factorsespecially inequality -are at play, others have emphasized other drivers, such as the role of party discipline (Canen, Kendall and Trebbi, 2021) and changes in the agenda (Lee, 2008).…”